Heuristics and Policies for Online Pickup and Delivery Problems
Oct 2012
The thesis focuses on the process of learning simple dispatch heuristics, and lays the foundations of a recommendation system able to rank such heuristics. Eight heuristics were implemented, observing different characteristics of the current fleet and orders. An artificial neural network that is trained on two hundred days of past data, and is supervised by schedules produced by an oracle, Indigo, which is a system able to produce suboptimal solutions to problem instances. We complement the quite promising results obtained with a discussion on future additions and improvements such as channel fleet management, traffic consideration, and learning hyper-heuristics to control simple rule sequences.
